{"title":"On Kolmogorov's superpositions and Boolean functions","authors":"Valeriu Beiu","doi":"10.1109/SBRN.1998.730994","DOIUrl":"https://doi.org/10.1109/SBRN.1998.730994","url":null,"abstract":"The paper overviews results dealing with the approximation capabilities of neural networks, as well as bounds on the size of threshold gate circuits. Based on an explicit numerical (i.e., constructive) algorithm for Kolmogorov's superpositions we show that for obtaining minimum size neural networks for implementing any Boolean function, the activation function of the neurons is the identity function. Since classical AND-OR implementations, as well as threshold gate implementations which require exponential size (in the worst case), it follows that size-optimal solutions for implementing arbitrary Boolean functions require analog circuitry. Conclusions and several comments on the required precision are presented.","PeriodicalId":401282,"journal":{"name":"Proceedings 5th Brazilian Symposium on Neural Networks (Cat. No.98EX209)","volume":"59 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1998-12-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127228397","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
A. Romariz, P. Ferreira, J. Campêlo, M. L. Graciano, O. R. Maia, J. C. Costa
{"title":"Realization of a mixed-mode neural coprocessor for signal processing","authors":"A. Romariz, P. Ferreira, J. Campêlo, M. L. Graciano, O. R. Maia, J. C. Costa","doi":"10.1109/SBRN.1998.731023","DOIUrl":"https://doi.org/10.1109/SBRN.1998.731023","url":null,"abstract":"A hybrid architecture for neural coprocessing is presented. A fixed set of analog multipliers and capacitors (analog memory) emulates multilayer perceptrons through digitally-controlled multiplexing. Parallelism is partially preserved, then, without direct analog implementation of the whole structure. Details of system VLSI implementation are given, along with simulation results that validate system cells design.","PeriodicalId":401282,"journal":{"name":"Proceedings 5th Brazilian Symposium on Neural Networks (Cat. No.98EX209)","volume":"47 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1998-12-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128437255","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"A realistic computer simulation of primary somatosensory cortex replicating static properties of topographic organization","authors":"M. Mazza, A. Roque-da-Silva","doi":"10.1109/SBRN.1998.731020","DOIUrl":"https://doi.org/10.1109/SBRN.1998.731020","url":null,"abstract":"A model of the somatosensory system has been constructed with the neuro-simulator GENESIS as a network of 1,024 pyramidal cells and 512 baskets cells connected to 512 tactile receptors representing the hand surface reproducing processes of formation and maintenance of somatotopic maps. The model presents results such as variability in the shapes and sizes of the areas of cortical representation, values of cortical magnifications in agreement with experimental findings and linear decay of receptive fields overlap as a function of cortical distance between recording sites in normal conditions.","PeriodicalId":401282,"journal":{"name":"Proceedings 5th Brazilian Symposium on Neural Networks (Cat. No.98EX209)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1998-12-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130654684","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Unsupervised neural network learning for blind sources separation","authors":"H. Szu, C. Hsu","doi":"10.1109/SBRN.1998.730990","DOIUrl":"https://doi.org/10.1109/SBRN.1998.730990","url":null,"abstract":"Review of independent component analyses (ICA) and blind sources separation (BSS) employing in terms of unsupervised neural networks technology are given. For example, imagery features occurring in human visual systems are the continuing reduction of redundancy towards the \"sparse edge maps\". When edges are multiplying together as the vector inner product they result in almost zero, namely pseudo-orthogonal ICA. This fact has been derived from the first principle of artificial neural networks using the maximum entropy information-theoretical formalism by Bell and Sejnowski (1996). We explore the blind de-mixing condition for more than two objects using two sensor measurement. We design two smart cameras with short term working memory to do better image de-mixing of more than two objects. We consider channel communication application that we can efficiently mix four images using matrices [A/sub 0/] and [A/sub 1/] to send through two channels.","PeriodicalId":401282,"journal":{"name":"Proceedings 5th Brazilian Symposium on Neural Networks (Cat. No.98EX209)","volume":"38 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1998-12-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121544745","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Extracting rules from feedforward Boolean neural networks","authors":"Teresa B Ludermir","doi":"10.1109/SBRN.1998.730995","DOIUrl":"https://doi.org/10.1109/SBRN.1998.730995","url":null,"abstract":"A method to extract rules from a feedforward Boolean neural networks (BNN) is described. We argue that rule extraction in BNN is more feasible and more natural than in other artificial neural networks models. This technique allows the understanding of how the neural networks reach a solution of a problem. A straight forward application of rule extraction from neural networks is the design of expert systems.","PeriodicalId":401282,"journal":{"name":"Proceedings 5th Brazilian Symposium on Neural Networks (Cat. No.98EX209)","volume":"18 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1998-12-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134531909","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"A new proposal for implementation of competitive neural networks in analog hardware","authors":"P. Engel, R.F. Molz","doi":"10.1109/SBRN.1998.731024","DOIUrl":"https://doi.org/10.1109/SBRN.1998.731024","url":null,"abstract":"This paper proposes the basic structure of a competitive neural network (CNN), based on analog IC techniques, for hardware implementation. This leads to a more compact project and allows real time processing. It is shown that the fundamental equations that describe the behavior of competitive neural networks possess a relationship with some basic electronic components. This fact allows the direct implementation of CNN with these electronic components. Initially the behavior of the fundamental equations of this type of neural networks is studied by means of software simulations. This behavior is then compared with the one obtained through electric simulations of the equivalent circuits originated from these fundamental equations. Simulations show that the most important features of the CNN are obtained with the presented implementation. Finally, a typical application is presented in the area of pattern clustering using synaptic weights to demonstrate an implementation using the techniques described.","PeriodicalId":401282,"journal":{"name":"Proceedings 5th Brazilian Symposium on Neural Networks (Cat. No.98EX209)","volume":"7 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1998-12-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132613586","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Fault detection and isolation in robotic manipulators and the radial basis function network trained by the Kohonen's self-organizing map","authors":"R. Tinós, M. Terra","doi":"10.1109/SBRN.1998.730999","DOIUrl":"https://doi.org/10.1109/SBRN.1998.730999","url":null,"abstract":"In this work, artificial neural networks are employed in a fault detection and isolation scheme for robotic manipulators. Two networks are utilized: a multilayer perceptron is employed to reproduce the manipulator dynamical behavior, generating a residual vector that is classified by a radial basis function network, giving the fault isolation. Two methods are utilized to choose the radial unit centers in this network. The first method, forward selection, employs subset selection to choose the radial units from the training patterns. The second employs the Kohonen's self-organizing map to fix the radial unit centers in more interesting positions. Simulations employing a two link manipulator and the Puma 560 manipulator indicate that the second method gives a smaller generalization error.","PeriodicalId":401282,"journal":{"name":"Proceedings 5th Brazilian Symposium on Neural Networks (Cat. No.98EX209)","volume":"21 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1998-12-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131158458","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"A general approach for density estimation in the n-dimensional Boolean space","authors":"A. D. P. Braga, M. Costa","doi":"10.1109/SBRN.1998.731017","DOIUrl":"https://doi.org/10.1109/SBRN.1998.731017","url":null,"abstract":"Kanerva (1988) presented a model to estimate the intersection of two circles, or hyperspheres, in the n-dimensional Boolean space. The results were applied to show the properties of the sparse distributed memory model. Braga and Aleksander (1994) presented a more general model that is able to estimate any density in the Boolean space and is not restricted to the intersection problem. In both previous works, the densities are calculated in relation to two fixed points in the space. It is shown that the previous work of Braga and Aleksander can be extended to estimate densities in relation to more than two fixed points. With these results, intersections of three or more hyperspheres can be estimated. The results can be applied to provide a more accurate study of the properties of Boolean associative memories.","PeriodicalId":401282,"journal":{"name":"Proceedings 5th Brazilian Symposium on Neural Networks (Cat. No.98EX209)","volume":"10 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1998-12-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128833176","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Target recognition using evolutionary neural networks","authors":"Elson Felix Mendes Filho, A. Carvalho","doi":"10.1109/SBRN.1998.731035","DOIUrl":"https://doi.org/10.1109/SBRN.1998.731035","url":null,"abstract":"The work presented here is part of the SAPRI Project-System for Acquisition, Processing and Recognition of Images-which is being developed for the Brazilian Naval Force. This project aims the treatment of radar images, supplying the specification of an environment, which can provide safety sailing and traffic control. This system involves areas like neural networks, image processing and pattern recognition. This paper presents some preliminary results achieved by using an evolutionary approach to optimize MLP network architectures for the target recognition task, using different learning algorithms.","PeriodicalId":401282,"journal":{"name":"Proceedings 5th Brazilian Symposium on Neural Networks (Cat. No.98EX209)","volume":"51 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1998-12-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116027228","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"A neural classifier employing biased wavelets","authors":"R. Galvão, T. Yoneyama","doi":"10.1109/SBRN.1998.731004","DOIUrl":"https://doi.org/10.1109/SBRN.1998.731004","url":null,"abstract":"Wavelet neural networks (WNN) can be understood as neural structures which employ a wavelet layer to perform an adaptive feature extraction in the time-frequency domain. This paper aims at providing some new insight into this emerging field, discussing basic concepts involved and also detailing aspects of training and initialization. Two modifications to the basic training algorithms are also proposed, namely the introduction of a bias component in the wavelets and the adoption of a weight decay policy. For illustration, a WNN is employed in a problem of ECG segment classification.","PeriodicalId":401282,"journal":{"name":"Proceedings 5th Brazilian Symposium on Neural Networks (Cat. No.98EX209)","volume":"52 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1998-12-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124771722","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}